水表训练

http://caffe.berkeleyvision.org/gathered/examples/mnist.html

问题： lenet 输出的概率总有一个1 

解决方法：用softmax 前面的一层，然后归一化到0-1，好像这个问题还是解决不了。

其实我们需要解决两个问题：

A。输出概率

B. 去掉一些扫出来的明显不是数字的图片，不显示。

现在用前面的一层可以解决输出的概率的问题，但是因为输入任何一个图片就会输出一个0，1 那么如果输入的图片不是数字，那么还是会输出一个比较大的数字。并不能扔掉这些图片。

问一下GQ 这个问题。（回复：

 正样本的话最大值输出可以在6000以上，负样本暂时没超过3000 ，暂时先做个阈值 把小的去掉，着实不是一个处理的好方法）

我们主要是训练了三个网络：输出都是0,1 

1.用最新的caffe-master 下面 examples 下面的mnist。lenet 的prototxt 是这样的。

name: "LeNet"
input: "data"
input_shape {
  dim: 1
  dim: 1
  dim: 28
  dim: 28
}
layer {
  name: "conv1"
  type: "Convolution"
  bottom: "data"
  top: "conv1"
  param {
    lr_mult: 1
  }
  param {
    lr_mult: 2
  }
  convolution_param {
    num_output: 20
    kernel_size: 5
    stride: 1
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "pool1"
  type: "Pooling"
  bottom: "conv1"
  top: "pool1"
  pooling_param {
    pool: MAX
    kernel_size: 2
    stride: 2
  }
}
layer {
  name: "conv2"
  type: "Convolution"
  bottom: "pool1"
  top: "conv2"
  param {
    lr_mult: 1
  }
  param {
    lr_mult: 2
  }
  convolution_param {
    num_output: 50
    kernel_size: 5
    stride: 1
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "pool2"
  type: "Pooling"
  bottom: "conv2"
  top: "pool2"
  pooling_param {
    pool: MAX
    kernel_size: 2
    stride: 2
  }
}
layer {
  name: "ip1"
  type: "InnerProduct"
  bottom: "pool2"
  top: "ip1"
  param {
    lr_mult: 1
  }
  param {
    lr_mult: 2
  }
  inner_product_param {
    num_output: 500
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "relu1"
  type: "ReLU"
  bottom: "ip1"
  top: "ip1"
}
layer {
  name: "ip2"
  type: "InnerProduct"
  bottom: "ip1"
  top: "ip2"
  param {
    lr_mult: 1
  }
  param {
    lr_mult: 2
  }
  inner_product_param {
    num_output: 20
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "prob"
  type: "Softmax"
  bottom: "ip2"
  top: "prob"
}

2. 在网上找的一个mnist的配置文件。输入图片的尺寸是32 *32的

name: "LeNet"
input: "data"
input_dim: 1
input_dim: 1
input_dim: 32
input_dim: 32
layer {
  name: "conv1"
  type: "Convolution"
  bottom: "data"
  top: "conv1"
  param {
    lr_mult: 1
  }
  param {
    lr_mult: 2
  }
  convolution_param {
    num_output: 6
    kernel_size: 5
    stride: 1
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "pool1"
  type: "Pooling"
  bottom: "conv1"
  top: "pool1"
  pooling_param {
    pool: MAX
    kernel_size: 2
    stride: 2
  }
}
layer {
  name: "conv2"
  type: "Convolution"
  bottom: "pool1"
  top: "conv2"
  param {
    lr_mult: 1
  }
  param {
    lr_mult: 2
  }
  convolution_param {
    num_output: 16
    kernel_size: 10
    stride: 1
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "ip1"
  type: "InnerProduct"
  bottom: "conv2"
  top: "ip1"
  param {
    lr_mult: 1
  }
  param {
    lr_mult: 2
  }
  inner_product_param {
    num_output: 120
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "relu1"
  type: "ReLU"
  bottom: "ip1"
  top: "ip1"
}
layer {
  name: "ip2"
  type: "InnerProduct"
  bottom: "ip1"
  top: "ip2"
  param {
    lr_mult: 1
  }
  param {
    lr_mult: 2
  }
  inner_product_param {
    num_output: 84
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "relu2"
  type: "ReLU"
  bottom: "ip2"
  top: "ip2"
}
layer {
  name: "ip3"
  type: "InnerProduct"
  bottom: "ip2"
  top: "ip3"
  param {
    lr_mult: 1
  }
  param {
    lr_mult: 2
  }
  inner_product_param {
    num_output: 20
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "prob"
  type: "Softmax"
  bottom: "ip3"
  top: "prob"
}

3. 加了均值之后的mnist为lenet2.

需要注意的问题：

A. 在用matlab 提特征的时候我们要吧prototxt 里面的数字

input_dim: 1
input_dim: 1
input_dim: 32
input_dim: 32

B. matlab 提特征的时候，要提那一层的特征把之后的层全去掉。比如我们想要softmax 之前层的特征，要把softmax 这一层全去掉。